DOCSLIB.ORG

Collaborative Software Performance Engineering for Enterprise Applications

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Collaborative Software Performance Engineering for Enterprise Applications Proceedings of the 50th Hawaii International Conference on System Sciences | 2017 Collaborative Software Performance Engineering for Enterprise Applications Hendrik Müller, Sascha Bosse, Markus Wirth, Klaus Turowski Faculty of Computer Science, Otto-von-Guericke-University, Magdeburg, Germany {hendrik.mueller, sascha.bosse, markus.wirth, klaus.turowski}@ovgu.de Abstract address these challenges from a software engineering perspective under the term DevOps [2], [3]. In the domain of enterprise applications, Corresponding research activities aim at increased organizations usually implement third-party standard flexibility through shorter release cycles in order to software components in order to save costs. Hence, support frequently changing business processes. application performance monitoring activities Therefore, DevOps enables a culture, practices and constantly produce log entries that are comparable to automation that support fast, efficient and reliable a certain extent, holding the potential for valuable software delivery [4]. However, especially for collaboration across organizational borders. Taking enterprise applications, IT departments usually make advantage of this fact, we propose a collaborative use of existing standard software components instead knowledge base, aimed to support decisions of of developing solutions entirely in-house [5]. In such performance engineering activities, carried out cases, activities referred to as Dev, rather include during early design phases of planned enterprise requirements engineering, architectural design, applications. To verify our assumption of cross- customization, testing, performance-tuning and organizational comparability, machine learning deployment [6]. Here, key service levels typically algorithms were trained on monitoring logs of 18,927 include performance objectives, expressed in terms of standard application instances productively running average response times, throughput or latency, in at different organizations around the globe. Using order to ensure deployment options that are aligned random forests, we were able to predict the mean with actual business requirements. Since enterprise response time for selected standard business applications support vital corporate business transactions with a mean relative error of 23.19 functions, their performance is critical for success of percent. Hence, the approach combines benefits of business tasks and must never be degraded existing measurement-based and model-based significantly [7]. As corporations tend to grow performance prediction techniques, leading to depending on an increased customer base, new competitive advantages, enabled by inter- product releases, new divisions and acquisitions [6], organizational collaboration. enterprise application performance needs to be monitored and managed continuously and proactively [8]. In order to evaluate and test software performance in early design stages, development 1. Introduction teams carry out software performance engineering (SPE) activities [9]. In addition, operations teams As of today, development and operations conduct application performance monitoring (APM) activities are not integrated tightly enough within one [10] to control the status of running systems. In a organization [1] and in most cases do not collaborate recent technical report, [1] summarizes that the across organizations, although similar and challenges of these two performance domains are comparable standard software components are often considered independently from each other. consistently deployed. The lack of integration and the Hence, they identified the interoperability between distribution of responsibilities across organizational tools and techniques of SPE and APM as a key units lead to increasing difficulties in accessing success factor in order to support a level of technical information that is required to deploy, operate and and organizational flexibility that is required for maintain applications cost-effectively and in constant DevOps. According to this review, performance data alignment to agreed service levels [1]. As a collected during operations provides various insights consequence, the end-user experience may suffer, for development teams and replaces assumptions e.g., from poor software performance caused by with knowledge. However, managing performance at inadequate capacity planning. Recent approaches design-time (SPE) or at run-time (APM) leads to URI: http://hdl.handle.net/10125/41196 ISBN: 978-0-9981331-0-2 CC-BY-NC-ND 391 different challenges since, for instance, more degrees than 18,000 productively running application of freedom exist at design-time for the system instances of a global market leading enterprise architecture [11]. In both domains, application resource planning software. Thus, the approach performance can be evaluated using either integrates data from different sources and utilizes measurement-based or model-based techniques [1]. methods and technologies related to the field of big Depending on the application life cycle, data and statistical analyses. measurement-based performance evaluation is Our approach combines the accuracy of real conducted during regular operations or as part of quality assurance systems with the cost-effectiveness stress tests or performance unit tests inside quality of performance models and data analysis. Hence, we assurance systems [12]. While model-based combine the advantages of model-based (applicable approaches can be used at design-time, at design-time) and measurement-based prediction parametrization is often difficult and involves expert (high accuracy of results) while not suffering from knowledge which may lead to low credibility of the their disadvantages (difficult parameterization, obtained results. Measurement-based approaches, on required expert knowledge, need for an implemented the other hand, require (parts of) an implemented system and sufficient training data size in operation). system and are normally limited in terms of the The following research questions narrow down configuration alternatives to be considered [11] our particular interest addressed in this paper: Do which means these are more suitable in the operation performance-related monitoring logs of enterprise phase. The advantages of both approaches could be standard software contain information that can be exploited if previously conducted measurements and extracted on a global scale in order to serve as a derived knowledge are made available for the design valuable input for new software rollouts? How of new landscapes. accurate are performance models that have been Gartner summarizes the efforts to integrate trained on mass data from different environments but monitoring data from different sources under the term similar software? Which model types are most IT operations analytics [13]. As an example, in [8], a suitable for log-based performance prediction? Our knowledge base is trained to collect performance research methodology follows the design science profiles of services in a SOA environment. These paradigm [14], where the developed artefact efforts aim at the integration of SPE and APM in a comprises the knowledge base including performance single organization (intra-organizational). In order to models that we evaluate in terms of accuracy and utilize knowledge for different organizations, the applicability in multiple iterations. specifics of this organization have to be considered in Therefore, the remainder of this paper is the applied performance models. However, in a structured as follows: In Section 2, we provide an standard application environment there is no need for overview of the state of the art related to model-based individual performance models and monitoring data and measurement-based software prediction. We can potentially be normalized and aggregated across conclude that both approaches have their several systems. Thus, knowledge that has been disadvantages. A knowledge base is proposed in applied to deploy existing landscapes can be Section 3 to combine the advantages of model-based extracted and applied to new scenarios, saving costs and measurement-based prediction as a concept and and time. Furthermore, the risk of errors, for as a technical architecture. In the evaluation, instance, by a service provider that conducts presented in Section 4, a prototype of the knowledge benchmarks, is avoided. To support intra- and inter- base is implemented and applied to monitoring data organizational collaboration for performance of nearly 19,000 business application systems. The activities during design and operations phase, we paper concludes with Section 5 in which the key follow an empirical approach that combines findings of the paper as well as possible future monitoring data from various instantiations of research activities are discussed. running enterprise applications and provides data- driven feedback for new software rollouts. 2. Software Performance Prediction Therefore, we investigate the comparability and applicability of this data to serve as an input for a domain-specific performance knowledge base which Software performance can be predicted using integrates different organizations that utilize the same either model-based or measurement-based standard software components. Within the knowledge techniques. In order to be able to classify our base, we utilize machine learning algorithms to build approach, this section summarizes the state of the art performance prediction models that we train using for
Load more

Recommended publications
  • A Survey of Collaborative Tools in Software Development
    1 INTRODUCTION A Survey of Collaborative Tools in Software Development Anita Sarma Institute for Software Research Donald Bren School of Information and Computer Sciences University of California, Irvine [email protected] ISR Technical Report # UCI-ISR-05-3 March 22, 2005 1 Introduction Collaboration is at the heart of software development. Virtually all software devel- opment requires collaboration among developers within and outside their project teams, to achieve a common objective. It has in fact been shown that about 70% of a software engineer’s time is spent on collaborative activities [219]. Indeed, col- laboration in software development has been studied by researchers in the fields of Software Engineering and Computer Supported Cooperative Work (CSCW) since the 1980s and has produced a wide range of collaborative tools. Enabling software developers to collaborate effectively and effortlessly is a difficult task. The collaboration needs of the team depend to a large extent on environmental factors such as, the organizational structure of the team, the domain for which the software is produced, the product structure, and individual team members. Accordingly, research in collaborative development has produced a host of tools, each typically focussing on a different aspect of collaboration. Most teams have their favorite repertoire of tools that has been built from historical use. These tools may not always be the best suited for the team, but the team still uses them nevertheless as the inertia and cost of trying out new tools surpasses the benefits. A number of classification frameworks exist that can be used to classify collaborative tools. In addition to placing the various tools in context, developers can use these frameworks to select the right mix of tools fit for their needs.
    [Show full text]
  • Collaborative Education in a Virtual Learning Environment Hani Bani
    Collaborative Education in a Virtual Learning Environment Hani Bani-Salameh Department of Software Engineering, The Hashemite University, Zarqa 13115, Jordan E-mail: [email protected] Jafar Al-Gharaibeh Architecture Technology Corporation, Eden Prairie, MN 55344, United States E-mail: [email protected] Clinton Jeffery Department of Computer Science, University of Idaho, Moscow, ID 83843, USA E-mail: [email protected] Ziad A. Al-Sharif Department of Software Engineering, Jordan University of Sci. & Tech, Irbid, Jordan E-mail: [email protected] Abstract: For the last few decades, there was tremendous interest in harnessing 3D virtual environments for education and training. Multi-user game-like environments make use of non-player characters and quest activities in tutoring and training. This approach incorporates successful active learning and incremental progress. This article presents issues encountered while adapting the multi-user online game genre for educational virtual environments. In this endeavour, non-player characters play a central role in organizing and delivering educational content. Educational virtual environments call for additional kinds of extension beyond generic user content creation as delivered in some virtual worlds like SecondLife. Support for these environments will emphasize the creation of new activities. A set of libraries and techniques named Portable Non-player character tutors and Quests reduces the effort needed to develop and integrate educational non-player characters. These non-player characters are introduced in CVE, a platform for rapidly developing educational virtual environments. The Portable Non-player character tutors and Quests framework enables formatted web-based exercises, quizzes, and educational content to be incorporated into Copyright © 2016 Inderscience Enterprises Ltd.
    [Show full text]
  • Engineering of IT Management Automation Along Task Analysis, Loops, Function Allocation, Machine Capabilities
    Engineering of IT Management Automation along Task Analysis, Loops, Function Allocation, Machine Capabilities Dissertation an der Fakultat¨ fur¨ Mathematik, Informatik und Statistik der Ludwig-Maximilians-Universitat¨ Munchen¨ vorgelegt von Ralf Konig¨ Tag der Einreichung: 12. April 2010 Engineering of IT Management Automation along Task Analysis, Loops, Function Allocation, Machine Capabilities Dissertation an der Fakultat¨ fur¨ Mathematik, Informatik und Statistik der Ludwig-Maximilians-Universitat¨ Munchen¨ vorgelegt von Ralf Konig¨ Tag der Einreichung: 12. April 2010 Tag des Rigorosums: 26. April 2010 1. Berichterstatter: Prof. Dr. Heinz-Gerd Hegering, Ludwig-Maximilians-Universitat¨ Munchen¨ 2. Berichterstatter: Prof. Dr. Bernhard Neumair, Georg-August-Universitat¨ Gottingen¨ Abstract This thesis deals with the problem, that IT management automation projects are all tackled in a different manner with a different general approach and different resulting system architecture. It is a relevant problem for at least two reasons: 1) more and more IT resources with built-in or asso- ciated IT management automation systems are built today. It is inefficient to try to solve each on their own. And 2) doing so, reuse of knowledge between IT management automation systems, as well as reuse of knowledge from other domains is severely limited. While this worked with simple stand-alone remote monitoring and remote control facilities, automation of cognitive tasks will more and more profit from existing knowledge in domains such as artificial intelligence, statistics, control theory, and automated planning. A common structure also would ease integration and coupling of such systems, delegating cognitive partial tasks, and switching between commonly defined levels of automation. So far, this problem is only partly solved.
    [Show full text]
  • Invisible Crowds in Cyberspace
    Invisible Crowds in Cyberspace: Mapping the Social Structure of the Usenet1 Marc A. Smith University of California, Los Angeles Published in: Smith, Marc. “Invisible Crowds in Cyberspace: Measuring and Mapping the Social Structure of USENET” in Communities in Cyberspace: Perspectives on New Forms of Social Organization.. London, Routledge Press, 1999 SURVEYING SOCIAL CYBERSPACES The Usenet is a quintessential Internet social phenomenon: it is huge, global, anarchic and rapidly growing. It is also mostly invisible. Although, it is the largest example of a conferencing or discussion group system,2 the tools generally available to access it only display leaves and branches - chains of messages and responses. None present the trees and forest. With hundreds of thousands of new messages every day, it is impossible to try to read them all to get a sense of the entire place. As a result, an overview of activity in the Usenet has been difficult to assemble and many basic questions about its size, shape, structure and dynamics have gone unanswered. How big is the Usenet? How many people post? Where are they from? When and where do they post? How do groups vary from one another and over time? How many different kinds of groups are there? How many groups successfully thrive and how many die? What do the survivors have that the others lack? How do different social cyberspaces connect and fit together and form a larger ecology? There is no shortage of questions. But we lack a historical record of the transformations of social cyberspaces3 just at the point when network interaction media are being widely adopted.
    [Show full text]
  • Adopting a Collaborative Software Development Process in the Scientific Community
    Adopting a Collaborative Software Development Process in the Scientific Community NOAA's 32nd Climate Diagnostics and Prediction Workshop (CDPW) Mike Charles DOC/NOAA/NWS/NCEP/CPC Motivation To advocate improved software development and project management techniques, including collaborative software development and the use of project management software. Who does this apply to? • Universities • Government • Private sector • Any software developers From now on, for the sake of brevity, "collaborative work" will include use of project management software. Disadvantages of Non- Collaborative Work Non-Collaborative Work A developer works alone • Work goes unchecked by others • If the owner leaves, nobody knows how the software works • There are no ideas contributed by others during development Non-Collaborative Work A developer doesn't log changes • Hard to revert to a previous version of the code • Can't remember what they did last week • Can't tell why a specific change was made Non-Collaborative Work A developer doesn't create documentation • Hard for others to adopt code • Hard for backup code owners to fix bugs and/or run code • Hard to share code with others Advantages of Collaborative Work Collaborative Work How should developers be working? • Collaboration - Working with others leads to more ideas and creativity, and more QCing of code • Tracking Changes - Using revision control software, tracking changes to code and keeping a log makes it easier to revert to older versions of the code, and keep track of work • Documentation - Creating
    [Show full text]
  • WHY USE a WIKI? an Introduction to the Latest Online Publishing Format
    WHY USE A WIKI? An Introduction to the Latest Online Publishing Format A WebWorks.com White Paper Author: Alan J. Porter VP-Operations WebWorks.com a brand of Quadralay Corporation [email protected] WW_WP0309_WIKIpub © 2009 – Quadralay Corporation. All rights reserved. NOTE: Please feel free to redistribute this white paper to anyone you feel may benefit. If you would like an electronic copy for distribution, just send an e-mail to [email protected] CONTENTS Overview................................................................................................................................ 2 What is a Wiki? ...................................................................................................................... 2 Open Editing = Collaborative Authoring .................................................................................. 3 Wikis in More Detail................................................................................................................ 3 Wikis Are Everywhere ............................................................................................................ 4 Why Use a Wiki...................................................................................................................... 5 Getting People to Use Wikis ................................................................................................... 8 Populating the Wiki................................................................................................................. 9 WebWorks ePublisher and Wikis
    [Show full text]
  • Personalizing the Collaborative Learning Environment with Pictures
    Personalizing the Collaborative Learning Environment with Pictures Brian Mackie OMIS Department, Northern Illinois University DeKalb, IL 60115, USA [email protected] and Charletta F. Gutierrez ABF Department, Northeastern Illinois University 5500 N. St. Louis Avenue Chicago, IL 60625, USA [email protected] ABSTRACT that are geographically and time-dispersed [3]. Making the most efficient use of time is critical in today’s fast-paced world. The Internet and Web-based technologies, as well as rapid Collaborative software can expedite the ability of groups to globalization, are changing the way businesses communicate. develop a shared understanding of a problem or task and move Continuous progress in Information Technology (IT) enables towards a solution. According to [4], new technology should be effective and efficient communication, particularly with the use developed in such a way that it enhances the existing culture. of collaborative systems. Such systems have many different That is, such technology should be developed from within the types of interfaces and attributes, and one such attribute is the use organization, rather than being pushed down from upper of visuals. This research assesses the usefulness of participant management [4]. Technologies should not just adapt to the pictures in a collaborative exchange. To evaluate the usefulness existing culture, but should help the organization change the of such pictures, participants were asked a series of questions existing norms; otherwise such technologies are likely to fail regarding the use of pictures in CAMS, a collaborative [5],[6]. To be effective, these technologies must do more than environment. The results suggest that, in a collaborative setting, automate existing practices; they should improve processes not the use of pictures is valuable in enhancing a “sense of only in the business organization but also in educational settings community,” particularly in cases where participants have not [6],[7].
    [Show full text]
  • Wikis As a Tool for Collaborative Language Learning: Implications for Literacy, Language Education and Multilingualism
    ISSN 2335-2019 (Print), ISSN 2335-2027 (Online) Darnioji daugiakalbystė | Sustainable Multilingualism | 5/2014 http://dx.doi.org/10.7220/2335-2027.5.8 Selami Aydin Balikesir University, Turkey WIKIS AS A TOOL FOR COLLABORATIVE LANGUAGE LEARNING: IMPLICATIONS FOR LITERACY, LANGUAGE EDUCATION AND MULTILINGUALISM Summary. This study presents a review of the literature concerning the use of wikis as a tool for collaborative learning in the second language acquisition and foreign language learning process, as research on the use of wikis is relatively new. The study first introduces the theoretical background behind the use of wikis in the mentioned processes. Then, it reviews the studies concerning wikis as a tool for collaborative language learning. The study concludes that the use of wikis improves basic language skills, encourages positive perceptions of learning, increases motivation, provides opportunities to practice autonomy, and enhances inquiry learning and critical thinking, although they appear to have no effect on cultural proficiency. More importantly, Wikis appear to be an advantageous tool for improving writing, and may have additional benefits in developing literacy in multilingual contexts where access to minor languages in the educational process is limited. The paper ends with practical recommendations for teachers and researchers. Keywords: collaborative learning; collaboration; foreign language learning; learner autonomy; second language learning. Introduction Within the contexts of second language acquisition and foreign language learning, wikis constitute a quickly emerging and popular learning tool. Because research on their efficiency is relatively new, wikis’ efficacy in the language acquisition and learning processes remains an open question. This study aims to review various scientific studies on the use of wikis as a tool for collaborative learning and their role within second language acquisition and foreign language learning contexts.
    [Show full text]
  • Enterprise Collaboration & Social Software
    Enterprise Collaboration & Social Software June 2013 INDUSTRY REPORT INSIDE THIS ISSUE Enterprise Collaboration & Social Software 1. Introduction INTRODUCTION 2. Market Trends This report focuses on technologies for collaboration and socialization within the enterprise. A number of forces are currently playing out in the enterprise IT 3. Competitive Landscape environment that are creating an inflection in the adoption and deployment of social and collaboration technologies. This significant uptrend has provided strong 4. M&A Activity growth for the sector and is driving a substantial amount of M&A and investment activity. This report includes a review of the recent M&A and private investing 5. Private Financings activities in enterprise social and collaboration software, particularly within the areas of group collaboration & workspaces, private social platforms, project and 6. Valution Trends social task management, event scheduling, web collaboration, white boarding & diagramming, and other related technologies. We have also profiled about 50 emerging private players in these subcategories to provide an overview of the 7. Emerging Private Companies breadth and diversity of the players targeting this sector. OVERVIEW Socialization and collaboration technologies are currently reshaping the established enterprise collaboration market as well as creating whole new categories of offerings, especially around private social platforms. In addition, many other enterprise applications such as CRM and unified communications are heavily transformed through the incorporation of new technologies including group messaging & activity feeds, document collaboration, and analytics. Much of this change is being driven by the consumerization of IT and the incorporation of social technologies. As businesses look to leverage the benefits of improved “connecting” and “network building” that employees have experienced with Facebook and other social solutions, a convergence is occurring between the enterprise social software and collaboration markets.
    [Show full text]
  • The Dos and Don'ts of Videoconferencing in Higher
    The Dos and Don’ts of Videoconferencing in Higher Education HUSAT Research Institute Loughborough University of Technology Lindsey Butters Anne Clarke Tim Hewson Sue Pomfrett Contents Acknowledgements .................................................................................................................1 Introduction .............................................................................................................................3 How to use this report ..............................................................................................................3 Chapter 1 Videoconferencing in Higher Education — How to get it right ...................................5 Structure of this chapter ...............................................................................................5 Part 1 — Subject sections ............................................................................................6 Uses of videoconferencing, videoconferencing systems, the environment, funding, management Part 2 — Where are you now? ......................................................................................17 Guidance to individual users or service providers Chapter 2 Videoconferencing Services — What is Available .....................................................30 Structure of this chapter ...............................................................................................30 Overview of currently available services .......................................................................30 Broadcasting
    [Show full text]
  • The Role of Collaborative Software and Decision Support Systems in the Smarter Cities
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Research Papers in Economics THE ROLE OF COLLABORATIVE SOFTWARE AND DECISION SUPPORT SYSTEMS IN THE SMARTER CITIES Virgil Chichernea1 Abstract The transition from the traditional city to the smart city is made by supported efforts regarding the achievement of a more steady, more efficient, more responsible city, through convergent strategies that deal with Smart Transportation Systems, Energy and Utilities Management, Water Management, Smart Public Safety, Healthcare Systems, Environmental Management, Educational Systems, Telecommunications (ITC Support),etc. and Positive Thinking. Service Oriented Architecture (SOA) meets the customers’ needs and the administration, the management of data, information, knowledge and decisions through Collaborative Systems and Decision Support Systems have a major impact both at the level of the smart city and the level of subsystems/services, and the information technology within smart cities becomes a major direction of research in the field of ITC. Keywords: Smart City, Collaborative Systems, Decision Support Systems (DSS), Service Oriented Architecture (SOA), Portal technology Introduction: Collaborative software (also referred to as groupware, workgroup support systems or simply group support systems) is computer software designed to help people involved in a common task achieve their goals. It is usually associated with individuals not physically co-located, but instead working together across an internet connection. It can also include remote access storage systems for archiving common use data files that can be accessed, modified and retrieved by the distributed workgroup members. Collaborative software is a broad concept that greatly overlaps with Computer-supported cooperative work (CSCW).
    [Show full text]
  • Proquest Dissertations
    Verification and Validation of UML and SysML Based Systems Engineering Design Models YOSR JARRAYA A THESIS IN The Department of Electrical and Computer Engineering Presented in Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy Concordia University Montréal, Québec, Canada April 2010 Yosr Jarraya, 2010 Library and Archives Bibliothèque et 1*1 Canada Archives Canada Published Heritage Direction du Branch Patrimoine de l'édition 395 Wellington Street 395, rue Wellington OttawaONK1A0N4 Ottawa ON K1A 0N4 Canada Canada Yourfìle Votre référence ISBN: 978-0-494-67341-6 Our file Notre référence ISBN: 978-0-494-67341-6 NOTICE: AVIS: The author has granted a non- L'auteur a accordé une licence non exclusive exclusive license allowing Library and permettant à la Bibliothèque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par télécommunication ou par l'Internet, prêter, telecommunication or on the Internet, distribuer et vendre des thèses partout dans le loan, distribute and sell theses monde, à des fins commerciales ou autres, sur worldwide, for commercial or non- support microforme, papier, électronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats. The author retains copyright L'auteur conserve la propriété du droit d'auteur ownership and moral rights in this et des droits moraux qui protège cette thèse. Ni thesis. Neither the thesis nor la thèse ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent être imprimés ou autrement printed or otherwise reproduced reproduits sans son autorisation.
    [Show full text]